peplexicon

Peplexicon Lab · Start here

What is a peptide, really?

You’ve seen the word on weight-loss drugs, supplements, and skin creams. Here’s what peptides are, how they’re made, how they work in the body, and how anyone knows if they help. Six short stops. No science background needed.

Choose your depth

The essentials in plain language. Your reading level. The same underlying facts.

Every picture here is simplified, and says so. Want more detail? Switch to Learning or Scientific at any time. The facts stay the same.

01BuildWhat a peptide is02DiscoverHow peptides are found03MakeHow peptides are made04TestHow a batch is checked05ActHow peptides work in the body06ProveHow we learn if it helps
One peptide, six benches.The open ring: questions still being studied.

01Build · What a peptide is

Beads on a string.

A peptide is a short chain of building blocks called amino acids. There are 20 kinds, and their order decides what the peptide does—like letters spelling a word. Pick a real one, or build your own.

Chain builderA simplified picture · not to scale

Setting up the 3D bench…

The gold bridge locks block 1 to block 6, closing a ring.

Try a real one

Or tap building blocks to make your own chain. The colours group blocks that behave alike.

Water-avoiding
Ring-shaped
Water-loving
Positive
Negative
Shape-setters
Building blocks
98 links between them
Kind
Peptideup to 40 blocks
Electric charge
Slightly positivein the body
Mixes with water
Fairly well

02Discover · How peptides are found

Where do new peptides come from?

Some are copied from our own bodies. Some come from animal venom. Some are improved versions of natural ones. And some are found by testing billions of random chains to see which one grabs a target. Try that last one below.

  1. 01

    Copied from the body

    Many medicines start as hormones our bodies already make, like insulin.

    Insulin, 1921
  2. 02

    Found in nature

    Animal venoms are full of peptides. A Gila monster’s venom led to a diabetes drug; a sea snail’s venom led to a pain medicine.

    Exendin-4, 1992
  3. 03

    Improved by chemists

    Scientists change a few parts of a natural peptide so it lasts longer in the body.

    Semaglutide, 2017
  4. 04

    Found by searching billions

    Make billions of random peptides and keep the few that stick to a target. Try it below.

    Phage display, 1985
Fishing for a matchA simulation · each dot is one virus
  • Good match
  • Weak match
  • No match
  • Fake match (sticks to plastic)
Rinses
0of up to 5
Good matches
0.4%of the dots
Fake matches
1.1%stuck to plastic
Good matches after each rinseDark blue is good matches. The bar fills with them as you rinse.
Start0%Top number: good matches
The winners so farScientists read the genes of the survivors to learn which peptides they carry. Each letter is one building block.
PeptideHow it sticksShare
GRINFKGNo match0.13%
HWINRKANo match0.13%
QAFVIDNNo match0.13%
TEWSAVQNo match0.13%
MWHGNMFNo match0.13%

These sequences are made up for this simulation.

About 100 years of peptide medicine

How we got here

  1. Found in the body

    Insulin is isolated

    Banting and Best extract insulin from the pancreas. A 14-year-old is treated in January 1922—the first peptide-hormone therapy.

    Source
  2. Method

    Oxytocin is made in a lab

    Vincent du Vigneaud synthesizes oxytocin, the first polypeptide hormone built from scratch.

    Source
  3. Method

    A protein gets a sequence

    Frederick Sanger completes the amino-acid order of insulin, proving each protein has a defined sequence.

    Source
  4. Method

    Chains anchored to beads

    Bruce Merrifield builds peptides on a solid resin bead, turning weeks of work into repeatable cycles.

    Source
  5. Medicine

    Cells become factories

    Human insulin made by engineered bacteria (Humulin) is approved: the first medicine made with recombinant DNA.

    Source
  6. Method

    Libraries of billions

    George Smith shows viruses can display peptides on their surface, making it possible to screen enormous libraries.

    Source
  7. Found in nature

    A lizard’s venom

    Researchers isolate exendin-4 from Gila monster venom. It acts on the GLP-1 receptor.

    Source
  8. Medicine

    A cone snail’s toxin

    Ziconotide, based on a peptide from cone snail venom, is approved for severe chronic pain.

    Source
  9. Medicine

    Exenatide approved

    A synthetic copy of exendin-4 becomes the first GLP-1 receptor agonist approved in the US.

    Source
  10. Medicine

    Semaglutide approved

    An engineered GLP-1 lasting about a week is approved as a once-weekly injection.

    Source
  11. Regulation

    The 40–amino acid line

    Medicines longer than 40 amino acids, including insulin, transition to regulation as biologics in the US.

    Source

03Make · How peptides are made

Built one piece at a time.

A machine builds a peptide by adding one building block at a time, to millions of chains at once. Each step works almost every time—but “almost” adds up. Press Run and watch.

Peptide builderShows 42 of the millions of chains on a real bead
  1. Uncap
  2. Add a block
  3. Rinse
Steps
8to add 8 blocks
Perfect chains
92%no mistakes
With a mistake
8%missed at least one step
Longer chains, more mistakesShare of chains with no mistakes, by length. Small misses add up.
0%25%50%75%100%1102030405060Building blocks in the chain40: counts as a protein after this99.999.097.0You: 99 in 100Oxytocin: 92.3%

Two ways to make a peptide

Chemistry

Made by a machine

What you just watched. Good for short chains, and chemists can use building blocks nature doesn’t.

Biology

Grown by yeast or bacteria

GeneYeast or bacteriaChainCleanFinish

Scientists give microbes the instructions (a gene), and the microbes build the chain for them. Longer chains, like insulin, are usually made this way.

  • Human insulin made by bacteria was approved in 1982—the first medicine made this way.
  • Ozempic’s label says the main chain of semaglutide is made by yeast.

04Test · How a batch is checked

Is it really what the label says?

Before a medicine can be sold, labs check that it’s the right molecule, that it’s pure, and that it’s free of germs. Here are two of those checks, run on the batch you just made.

Purity checkSimplified · the tall peak is your peptide
11.912.914.015.116.117.218.3Time to come out of the machine (minutes)AmountOxytocin · 93%a mistakea mistake
The real thing
93%raw batch
Mistakes
7%chains missing a block
Weight checkIs it the right molecule?
Expected weight of Oxytocin
1,007.2
Measured
1,007.2
Result
Right molecule

A machine called a mass spectrometer weighs molecules. A chain missing one block would weigh less, so this check catches many mistakes. Weights are in daltons, a unit sized for molecules.

The checks a medicine must pass

Before a batch of an approved medicine can be sold, it has to pass every check on a list that regulators have reviewed. Here are four of them.

  1. Is it the right molecule?

    Weigh the molecules and compare with the expected weight.

    Checked above: yes
  2. How pure is it?

    Sort the batch and measure how much is the real thing.

    Checked above: 93%
  3. Is it free of bacterial toxins?

    A test that detects toxins from bacteria, which can cause fever if injected.

    Not shown here
  4. Is it free of germs?

    Grow samples to check that nothing living is in a product meant for injection.

    Not shown here

A clean test result tells you what’s in the vial—not whether it works or is safe for you. That takes the studies at the last stop. Products sold “for research use only” aren’t approved for people and don’t have to pass these checks.

05Act · How peptides work in the body

A key, a lock, and a clock.

A peptide works like a key. It fits a lock on a cell, called a receptor, and tells the cell to do something. But the body clears peptides out fast, so much of the cleverness in modern peptide medicines is making them last.

Locks on a cellA simplified picture
Outside the cellInside the cellBlue: the cell got the messageDots: peptide keys · Striped shapes: locks (receptors)
The key
Locks filled
50%4 of 8 shown
Switched on
50%sending a signal
How long it lastsSimplified · one dose · not a dosing guide
albumin (big blood protein)fatty tailAibThe scissors can’t cut herestartend31 building blocks0%50%100%1 min1 h1 day1 week4 weeksTime since the dose (log scale)half remainingGLP-1ExenatideLiraglutideSemaglutide
Half gone in
~1 week
Still in the blood
91%after 24 h
Vs. natural GLP-1
5,040×longer lasting
Protected from
the scissors · the kidneys

06Prove · How we learn if it helps

How do we know if it works?

Something that works in a dish or in mice may not work in people. The only way to know is to test it in people, step by step, with a fair comparison. Try running a fair test yourself.

The steps to proofBased on FDA’s guide for patients

Big tests (Phase 3)

Does it work better than a dummy treatment or current care?

Who or what is tested
300 to 3,000 people with the condition
It can show
Confirm the benefit and the common side effects.
It can’t show
Catch very rare or very long-term harms.
On Peplexicon, labelled as
Fair tests in people

About 28 in 100 medicines at this step make it to the next one, by FDA’s estimate.

Of every 100 medicines tested in people
Start tests in people100
Reach medium tests70
Reach big tests23
Reach FDA reviewabout 6

Most never make it. And many more fail before they’re ever tested in people.

Semaglutide passed every step, including big tests in thousands of people, and was approved in 2017. See what the research found →

BPC-157 has mostly been tested in animals. It isn’t FDA-approved, and there’s no reliable evidence yet that it helps people.

FDA: The drug development process, Step 3: clinical research ↗

Fair-test simulatorSimulated people · not real data
Each dot is one person. Farther right means they improved more.-20-100+10+20+30Run the test to see the people in it.Each line is one test: the range its result could really be. Lines crossing 0 can’t rule out “no effect.”-100 · no difference+10+20Each test you run adds one line.
The truth
Result
—run a test
Looked like it worked
—tests so far
Chance to spot it
20%for a test this size

End of the tour

Now you know what to ask.

When you read about any peptide—here or anywhere else—ask three questions. Was it tested in people, or only in animals? Was it a fair test with a comparison group? Is it approved, or sold “for research use only”?

What this tour can’t tell you

  • These are simple pictures. Real molecules are too small to see, and real labs are messier.
  • Nothing here predicts what will happen to you. Bodies differ, and so do products.
  • This is not medical advice. Talk to a doctor or pharmacist about any medicine, including peptides.
Every source used in the lab (33)
  1. PubChem: Oxytocin (CID 439302)
  2. PubChem: Argipressin (CID 644077)
  3. UniProt P01275: Pro-glucagon, GLP-1(7-37)
  4. PubChem: Exenatide (CID 45588096)
  5. PubChem: BPC-157 (CID 9941957)
  6. PubChem: Aspartame
  7. Byetta (exenatide) prescribing information, DailyMed
  8. Drugs@FDA: Byetta, NDA 021773
  9. Victoza (liraglutide) prescribing information, DailyMed
  10. Drugs@FDA: Victoza, NDA 022341
  11. Ozempic (semaglutide) prescribing information, DailyMed
  12. Drugs@FDA: Ozempic, NDA 209637
  13. Rybelsus (oral semaglutide) prescribing information, DailyMed
  14. Holst JJ. The physiology of glucagon-like peptide 1. Physiol Rev. 2007
  15. Eng J et al. Isolation and characterization of exendin-4 from Heloderma suspectum venom. J Biol Chem. 1992
  16. Drugs@FDA: Prialt (ziconotide), NDA 021060
  17. Merrifield RB. Solid phase peptide synthesis. I. J Am Chem Soc. 1963
  18. Nobel Prize in Chemistry 1984: Bruce Merrifield
  19. Nobel Prize in Chemistry 1955: Vincent du Vigneaud
  20. Nobel Prize in Chemistry 1958: Frederick Sanger
  21. Smith GP. Filamentous fusion phage. Science. 1985
  22. Nobel Prize in Chemistry 2018: phage display
  23. Nobel Prize in Chemistry 2002: mass spectrometry of biomolecules
  24. Nobel Prize in Physiology or Medicine 1923: insulin
  25. Drugs@FDA: Humulin, application 018780
  26. FDA: “Deemed to be a license” provision of the BPCI Act
  27. FDA: The drug development process, Step 3: clinical research
  28. Wang L et al. Therapeutic peptides: current applications and future directions. Signal Transduct Target Ther. 2022
  29. Carpino LA, Han GY. The 9-fluorenylmethoxycarbonyl amino-protecting group. J Am Chem Soc. 1970
  30. Behrendt R, White P, Offer J. Advances in Fmoc solid-phase peptide synthesis. J Pept Sci. 2016
  31. FDA guidance: Pyrogen and endotoxins testing, questions and answers
  32. ExPASy: amino acid residue masses
  33. Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol. 1982